The interest for patient trajectory prediction, a sort of computer-aided medicine, has steadily increased with the pace of artificial intelligence innovation. Notwithstanding, the design of effective systems able to predict clinical outcomes based on the history of a patient is far from trivial. Works so far are based on neural architectures with low performance, especially when using low-cardinality datasets; alternatively, complex inference approaches are hard to reproduce and/or extrapolate as they are designed for very specific circumstances. We introduce LIG-Doctor, an artificial neural network architecture based on two Minimal Gated Recurrent Unit networks functioning in a bidirectional parallel manner, benefiting from temporal events both forward and backward. In comparison to state-of-the-art works, consistent improvements were achieved in prognosis prediction, as assessed with metrics Recall@k, Precision@k, F1-score, and AUC-ROC. Besides the detailed delineation of our architecture, a sequence of experiments is reported with insights that progressively guided design decisions to inspire future works on similar problems. Our results shall contribute to the improvement of computer-aided medicine and, more generally, to processes related to the design of neural network architectures.
Time-series forecasting is one of the most active research topics in artificial intelligence. It has the power to bring light to problems in several areas of knowledge, such as epidemiological studies, healthcare inference, and climate change analysis. Applications in real-world time series should consider two factors for achieving reliable predictions: modeling dynamic dependencies among multiple variables and adjusting the model's intrinsic hyperparameters. An open gap in the literature is that statistical and ensemble learning approaches systematically present lower predictive performance than deep learning methods. The existing applications consistently disregard the data sequence aspect entangled with multivariate data represented in more than one time series. Conversely, this work presents a novel neural network architecture for time-series forecasting that combines the power of graph evolution with deep recurrent learning on distinct data distributions, named after Recurrent Graph Evolution Neural Network (REGENN). The idea is to infer multiple multivariate relationships between co-occurring time-series by assuming that the temporal data depends not only on inner variables and intra-temporal relationships (i.e., observations from itself) but also on outer variables and inter-temporal relationships (i.e., observations from other-selves). An extensive set of experiments was conducted comparing REGENN with tens of ensemble methods and classical statistical ones. The results outperformed both statistical and ensemble-learning approaches, showing an improvement of 64.87% over the competing algorithms on the SARS-CoV-2 dataset of the renowned John Hopkins University for 188 countries simultaneously. For further validation, we tested our architecture in two other public datasets of different domains, the PhysioNet Computing in Cardiology Challenge 2012 and Brazilian Weather datasets. We also analyzed the Evolution Weights arising from the hidden layers of REGENN to describe how the variables of the dataset interact with each other; and, as a result of looking at inter and intra-temporal relationships simultaneously, we concluded that time-series forecasting is majorly improved if paying attention to how multiple multivariate data synchronously evolve.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.